IRC logs for #openrisc Thursday, 2015-04-23

--- Log opened Thu Apr 23 00:00:19 2015
stekernolofk: I wish you could refrain from being so politically incorrect ;)05:38
olofkI have a weird emacs problem. When I'm at the office, it can take up to a minute to launch, but when I'm at home it's more or less instant. I _think_ it's some kind of network problem, like slow DNS server or something, but haven't got a clue how to work around it.06:24
olofkstekern, dalias : Regarding the problems with musl, linux-headers and busybox. What is The Right Way (tm) to fix this? Could we solve it by only patching busybox, or will we need hacked kernel headers also?06:29
amsolofk: check what localhost resolves to06:34
olofkehhm... how to I do that?06:37
amscat /etc/hosts06:39
amshmm.. looks fine06:39
olofkCould it help perhaps to remove the ipv6 entry? Don't have ipv6 compiled at all06:39
olofkstrace produces like a mile of logs. Haven't investigated further06:40
amswell, i was thinking if it stoped where it hanged for a bit ..06:41
amsthat could lead to some stuff ..06:41
amstry emacs -Q? maybe your .init ..06:41
olofkyes, emacs -q helps a bit06:41
ams"a bit"?06:41
olofkIIRC it was a bit faster, but still a bit slow.06:42
olofkrecvfrom(4, "L\323\201\202\0\1\0\0\0\0\0\0\vsamanthafox\0\0\34\0\1", 2048, 0, {sa_family=AF_INET, sin_port=htons(53), sin_addr=inet_addr("")}, [16]) = 2906:43
olofkpoll([{fd=4, events=POLLIN}], 1, 494106:43
olofkIt hangs there for a while06:43
amshmm... what is
olofkI think that's YOUR MAMA!06:43
olofkOr the company nameserver06:44
olofkProbably the second alternative06:44
amsemacs -q is still slow though?06:44
olofkIt faster, but it takes a while after these lines06:45
olofkrecvfrom(4, "L\323\201\202\0\1\0\0\0\0\0\0\vsamanthafox\0\0\34\0\1", 2048, 0, {sa_family=AF_INET, sin_port=htons(53), sin_addr=inet_addr("")}, [16]) = 2906:45
olofkpoll([{fd=4, events=POLLIN}], 1, 494106:45
olofkpselect6(4, [3], [], NULL, {0, 999999094}, {NULL, 8}06:45
olofkThat one06:45
amsdoes it loop?06:45
amslike multiple pselects ..06:46
olofkThat might actually be some gtk thing. I get those when I move the mouse over the window06:46
olofkLooks like emacs -q also waits for the nameserver06:47
amshwat does M-x emacs-init-time say?06:47
amsmaybe the nameserver is slow?06:47
amswhat about the hostname of the machine?06:47
amsis it in /etc/hosts?06:47
amsi know that could cause issues ..06:47
olofkI'm still not totally ok that my text editor takes a minute to start because the DNS server is slow.06:47
amsresolving olofk-big-beard-machine to a ip06:48
amsolofk: well, it sounds .. kinda as if an incorrectly setup machine06:48
olofkShe's called samanthafox and she's using dynamic IP06:48
ams10.112.0.11 doesn't happen to be that ip?06:48
olofkNo matter how incorrect my network is set up. It *still* shouldn't affect the startup of my text editor06:48
amsof course it should, and it will.. emacs needs to know stuff06:49
olofkNope. .11 is the nameserver. I got .22306:49
olofkwhat stuff?06:49
amsand send it to the RMS organisation..06:49
amsolofk: your .. ~/.hidden folder06:49
olofkemacs-init-time say 1.2 seconds. That's so not true06:49
amsand those kitten pictures you send to stekern.06:50
olofkhey! Those were private06:50
amsso it works on another box, just fine, and here it does not.06:50
amsno clue man ..06:52
amsmaybe try it in batch mode and see if you can figure it out that way?06:54
olofkI found plenty of articles about this issue now06:55
olofkStill haven't found anyone who can tell me why the hell emacs need to know my hostname06:55
olofkfor example06:55
ams*** Emacs startup on GNU/Linux systems (and possibly other systems) is slow.olofk:06:57
amsThis can happen if the system is misconfigured and Emacs can't get the06:57
amsfull qualified domain name, FQDN.  You should have your FQDN in the06:57
amsThe way to set this up may vary on non-GNU systems.06:57
olofkYeah, I'm trying to figure out how to set up this FQDN thing06:59
amsin hosts?06:59
olofkI could try that07:00
olofkJust read that " Arnt Gulbrandsen <> says that # should NEVER be named with the name of the machine. It causes problems # for some (stupid) programs, irc and reputedly talk"07:00
amswell, you should atleast see that samantha resolves to whatever it should resolve to ..07:01
olofkShould the name and ip be like that, or should it be samanthafox ?07:02
amsgood one.07:03
amsi don't have anything in my hosts..07:03
olofkwhat does hostname and hostname -f say?07:04
amsfor both07:04
amsping hydrogenium works07:04
olofkhostname -f ==>  hostname: Host name lookup failure07:05
olofkhostname says samanthafo07:05
stekernams: you mean this?
amsmaybe fix the fqdn?07:09
amsstekern: no..07:09
olofk127.0.0.1       samanthafox in my /etc/hosts solved the problem07:18
amsmaybe check your dns server so that ghets fixed automatically?07:23
olofkok. I fixed it for now, but I probably need to wireshark the traffic to see what I'm sending to the dhcp server and what I get back07:38
daliasolofk, need to see the specific failure09:08
olofkdalias: Don't have the details now, but it was something about ethaddr_t being redeclared, and the headers from sabotage-linux fixes it10:09
daliasi thought most/all such issues were fixed on the busybox side now10:12
olofkdalias: I tried latest (0.22.3?)10:12
daliasit might just be in an applet i don't use10:12
olofktrue. If you have a known working busybox config, I could try to use that10:13
olofkBut from your answer I assume that it should be possible to fix this by just changing busybox10:14
olofk1.23.2 is the busybox version I tried btw10:24
olofkwow.. someone on the OpenCores forum is asking about some GCC changes he sees when upgrading to 4.9.2 from 3.4.411:15
Heshamolofk: How much is some?11:17
olofkI'm impressed that gcc 3.4.4 produces any code at all11:24
olofkI realized that IP-Xact isn't as good idea as I had hoped. Forgot that xml is an unordered mess that is almost as bad as binaries when it comes to version control11:58
olofkAt least when you let tools do the xml for you11:59
stekernmmm, it's a human readable format that's not readable by humans12:19
olofkI've just added an option to FuseSoc that I should have done years ago19:14
olofkFuseSoC now rebuilds the simulation models by default, instead of having to pass --force all the time19:15
olofkAnd --keep prevents rebuilding19:15
--- Log closed Fri Apr 24 00:00:20 2015

Generated by 2.15.2 by Marius Gedminas - find it at!