IRC logs for #openrisc Sunday, 2013-05-12

--- Log opened Sun May 12 00:00:55 2013
-!- hno` is now known as hno02:03
stekernis booting with rootfs over nfs supposed somehow broken on recent openrisc kernels/nfs servers, or is it just me that is doing something wrong?13:59
stekern-supposed14:00
stekernmounting the nfs works when booting with initramfs, so the server setu should be ok, i guess14:03
LoneTech_I don't know any specific reason that should be broken15:15
-!- LoneTech_ is now known as LoneTech15:15
--- Log closed Sun May 12 15:25:59 2013
--- Log opened Sun May 12 15:27:01 2013
-!- ServerMode/#openrisc [+ns] by rajaniemi.freenode.net15:27
-!- Irssi: #openrisc: Total of 1 nicks [1 ops, 0 halfops, 0 voices, 0 normal]15:27
-!- You're now known as 21WAAKUO315:29
-!- Irssi: Join to #openrisc was synced in 188 secs15:29
-!- ServerMode/#openrisc [-o 21WAAKUO3] by rajaniemi.freenode.net15:30
!rajaniemi.freenode.net *** Notice -- TS for #openrisc changed from 1368372421 to 130192679015:30
-!- ServerMode/#openrisc [+ct-s] by rajaniemi.freenode.net15:30
-!- ServerMode/#openrisc [+o juliusb] by rajaniemi.freenode.net15:30
-!- rajaniemi.freenode.net changed the topic of #openrisc to: Official OpenRISC Project channel | http://opencores.org/or1k | http://openrisc.net | wiki: http://opencores.org/or1k/OR1K:Community_portal | Mailing lists: http://lists.openrisc.net/ http://lists.opencores.org/ | forum: http://opencores.org/forum,OpenRISC | git: https://github.com/openrisc | IRC log: http://juliusbaxter.net/openrisc-irc/ | OpenCores in #opencores15:30
stekernbut, this is before busybox kicks in, in busybox i can mount the share15:30
stekernthat doesn't rule out nfs3/nfs4 issues though, I have enabled nfs4 support in the kernel though15:32
LoneTechstekern: no, sorry16:05
21WAAKUO3stekern: (zero delay slot penalty) you're right on having to have all branch targets being calculated in the fetch stage for zero branch/fetch delays to occur16:07
21WAAKUO3your branch prediction sounds interesting. It'd be very interesting to see the results of what branch strategy works best16:13
21WAAKUO3regarding data endianess, I think it'd be easy to swap the endianness, perhaps just in the LSU. I can't think of anything which manipulates the different bytes/shorts in the 32-bit word, except for maybe the sign extension instructions16:16
21WAAKUO3very odd... my name is somethign random - I didn't even check that16:18
-!- You're now known as juliusb16:18
stekernjuliusb: kind of guessed it was you hiding behind that odd nick =p16:28
tremHi all, I've got a little question, where is the latest code for ork1ksim ? (on svn  http://opencores.org/ocsvn/openrisc/openrisc/trunk/or1ksim   or on git https://github.com/openrisc/or1ksim)17:03
stekerntrem: I think the github one is the most up to date17:45
tremstekern: thanks17:45
tremgot an error on "make check"17:45
stekernjuliusb: I agree, it'll be interesting to check different branch prediction strategies, and I think it'd be cool to make the actual prediction part pretty modular, so you could go overboard and do some really fancy dynamic branch prediction, just for the kicks19:02
stekernbut right now, I think there are still more bugs in the pipeline control, I see some crashes that I've not seen before adding it19:03
stekernmost pipeline control changes are in the fetcher, so it's kind of expected to be a pita...19:03
stekernthis nfsboot is driving me crazy, it's not working with an old kernel and another nfs server neither21:08
stekernit actually seems the "board" isn't sending anything (apart from the dhcp request, that succeeds)21:09
stekernI bet this is something silly, like a typo in the ip-address or something, but I've double checked a million times21:11
stekernconsole=uart,mmio,0x90000000,115200 root=/dev/nfs nfsroot=192.168.255.100:/home/stefan/openrisc/nf\21:12
stekernsroot,rw,nolock ip=dhcp rw21:12
stekernis the kernel command line...21:13
--- Log closed Mon May 13 00:00:56 2013

Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!