| stekern | http://www.eetimes.com/electronics-blogs/other/4372270/Comparing-four-32-bit-soft-processor-cores?Ecosystem=programmable-logic | 09:45 |
|---|---|---|
| jeremybennett | stekern: Excellent - Clive Maxfield picking up on OpenRISC in EE Times. | 10:42 |
| stekern | yup, with the help of lobbying from our friend Sven-Åke ;) | 10:45 |
| jeremybennett | Of course - I've updated the tutorials entry on the Wiki, to give explicit credit to Sven-Åke | 10:53 |
| jeremybennett | I ought to update the Wikipedia entry as well. | 10:53 |
| jeremybennett | Wikipedia updated: http://en.wikipedia.org/wiki/OpenRISC | 11:33 |
| juliusb | stekern: sorry, been a few things on around here lately, have been monitoring the room but not had time to respond :( | 13:09 |
| juliusb | I have been building that binutils with the bld-all.sh script (so unisrc method) from the opencores SVN, and just specifying the path to that binutils dir | 13:10 |
| juliusb | so however that configured binutils is how I've been doing it | 13:10 |
| juliusb | yes, very cool to see sven ake's work | 13:12 |
| juliusb | so, long weekend comign up here in the UK | 13:12 |
| juliusb | I plan on buying a desk and doing some more openrisc hackery | 13:13 |
| stekern | "that binutils dir" = or1k-src binutils dir? | 13:13 |
| juliusb | it's horrible weather as usual | 13:13 |
| juliusb | umm, no, i just pointed it to my git checkout | 13:13 |
| juliusb | i haven't looked at the or1k-src stuff yet | 13:13 |
| stekern | buying a desk? don't you own a desk? | 13:13 |
| juliusb | but I presume pgavin has pushed the latest stuff | 13:14 |
| stekern | ah, ok I thought you had | 13:14 |
| juliusb | no desk yet, no, I moved house about a month ago and still haven't got one! hence not much FPGA action and so no testing of openocd stuff | 13:14 |
| stekern | I guess I have to ask pgavin about that then | 13:15 |
| juliusb | my last bit of work I pushed to pgavin's binutils tree on github and the idea was to put it on the opencores svn under the gnu-dev dir or whatever | 13:15 |
| juliusb | in fact, it was mostly gcc fixes I think, which got us to the gcc regression test results I documented on the wiki a few weeks ago | 13:15 |
| stekern | yeah, I saw that | 13:16 |
| stekern | I was mostly interested in binutils, wanted to try doing llvm+binutils without gcc as the middle man | 13:17 |
| stekern | turns out it's a bit of a hassle, you have to pipe the assembler output to as and then ld the output | 13:18 |
| stekern | using gcc as a shell for as and ld is a bit "cleaner" | 13:18 |
| juliusb | ah yea | 13:20 |
| juliusb | annoying | 13:20 |
| juliusb | so one of my projects last weekend was to fix up my or1ktrace libraryt hing | 13:20 |
| juliusb | i got it working using the latest binutils, nicely | 13:20 |
| juliusb | so that'l be handy for the mor1kx testbench stuff | 13:22 |
| juliusb | i'm hoping to address that this weekend, actually, neaten it all up ready for release | 13:22 |
| stekern | nice | 13:22 |
| stekern | will there be a release party? ;) | 13:27 |
| jemarch | hi | 13:32 |
| juliusb | haha, we'll see :) | 13:53 |
| pgavin | hello | 13:58 |
| juliusb | heya | 13:59 |
| pgavin | juliusb: are you here? | 13:59 |
| pgavin | ah | 13:59 |
| juliusb | yep | 13:59 |
| pgavin | hi | 13:59 |
| juliusb | :) | 13:59 |
| pgavin | :) | 13:59 |
| juliusb | hi | 13:59 |
| stekern | hello, just the guy I was waiting for ;) | 13:59 |
| juliusb | what's up? | 13:59 |
| pgavin | I wanted to ask you: do you know if there's a way to use or1ksim with the gcc testsuite? | 13:59 |
| juliusb | pgavin: ah yes - to run it directly against or1ksim instead of using GDB? | 14:00 |
| juliusb | or, rather, the gdb sim? | 14:00 |
| juliusb | that's what I've been doing | 14:00 |
| pgavin | ok, and my gcc tree works with that? | 14:00 |
| juliusb | so basically, it's doable as long as you can get or1ksim to return the the return code of the simulated program | 14:00 |
| pgavin | ok | 14:00 |
| juliusb | yep, that's how I ran all of the regression tests for the recent set of results I got | 14:00 |
| pgavin | right, that's what I thought | 14:00 |
| juliusb | and my patch I've been banging on about for or1ksim adds that feature to or1ksim | 14:00 |
| pgavin | ok | 14:00 |
| juliusb | but it's not committed yet because for the life of me I can't get that bloody tcl except junk to test the actual return code of or1ksim, as simple as it iseems | 14:01 |
| pgavin | hmm | 14:01 |
| juliusb | and i've sort of given up and have half a mind to just commit it without a test | 14:01 |
| juliusb | anyway | 14:01 |
| juliusb | i can post the patch and you can build or1ksim easily enough with this feature | 14:01 |
| juliusb | and then i've been using the gnu regression harness scripts from the opencores svn | 14:01 |
| pgavin | ok | 14:02 |
| juliusb | and just told it to call or1ksim instead of or32-elf-run or whatever | 14:02 |
| juliusb | and that appears to work fine | 14:02 |
| stekern | pgavin: I tried compiling or1k-src yesterday, but it gave up in ./gdb with a: configure: error: configuration or1k-unknown-elf is unsupported. | 14:02 |
| stekern | how are you supposed to build that thing? | 14:02 |
| juliusb | i can provide all of the details on the wiki tomorrow or perhaps tonight | 14:02 |
| pgavin | stekern: yeah, gdb isn't working yet :) | 14:02 |
| pgavin | stekern: you have to do --disable-gdb | 14:02 |
| pgavin | and --disable-sim | 14:03 |
| pgavin | stekern: I have a bunch of in-progress gdb stuff that I haven't pushed to the tree | 14:03 |
| stekern | ok, nice | 14:04 |
| pgavin | but I've had lots of other things to take care of, so I haven't done much on it in the last week or so | 14:04 |
| stekern | and thanks, I try rebuilding | 14:04 |
| pgavin | you can use --disable-[dirname] to keep it from building anything you don't care about, like tcl, tk, and so on | 14:05 |
| pgavin | it speeds up the build | 14:05 |
| stekern | ah, ok, thanks for the pointer | 14:05 |
| pgavin | but binutils, gas, ld, newlib should all work | 14:05 |
| pgavin | if they don't let me know :) | 14:05 |
| stekern | I have to configure with CC="gcc -fPIC" and CXX="g++ -fPIC" for it to compile on my machine too | 14:06 |
| pgavin | hmm | 14:07 |
| pgavin | this is what I use: --target=or1k-elf --enable-cgen-maint --disable-shared --disable-itcl --disable-tk --disable-tcl --disable-winsup --disable-libgui --disable-rda --disable-sid --disable-sim --disable-gdb --with-sysroot | 14:07 |
| pgavin | and if you haven't compiled gcc yet you need --disable-newlib --disable-libgloss | 14:07 |
| pgavin | once you have gcc compiled, you can go back and rebuild it with newlib enabled | 14:08 |
| stekern | ok, I'll try those options | 14:10 |
| stekern | hmm, now it complains about some guile.scm | 14:21 |
| pgavin | if you don't have guile installed, remove the --enable-cgen-maint option | 14:37 |
| pgavin | or just install guile :) | 14:37 |
| pgavin | but you shouldn't need the --enable-cgen-maint option unless you change the files under cpu/ | 14:37 |
| stekern | is some particular version guile version required? | 14:38 |
| pgavin | don't think so | 14:38 |
| stekern | yes, the removing --enable-cgen-maint works | 14:38 |
| pgavin | good :) | 14:38 |
| stekern | I installed, since it first complained about not finding it, but then it choked on guile.scm | 14:39 |
| stekern | +guile | 14:39 |
| stekern | but I'll settle for without --enable-cgen-maint for now ;) | 14:40 |
| pgavin | yeah, you shouldn't need it | 14:41 |
| pgavin | but I actually think you'd need guile-1.6 or better, since you asked | 14:41 |
| --- Log closed Fri May 04 14:47:49 2012 | ||
| --- Log opened Fri May 04 14:48:02 2012 | ||
| -!- Irssi: #openrisc: Total of 18 nicks [0 ops, 0 halfops, 0 voices, 18 normal] | 14:48 | |
| stekern | I installed guile-2.0 | 14:48 |
| -!- Irssi: Join to #openrisc was synced in 25 secs | 14:48 | |
| stekern | but maybe guile-1.6 is needed then | 14:48 |
| pgavin | guile-1.8 workds | 14:49 |
| pgavin | works* | 14:49 |
| stekern | ok, I can look into that closer (if I ever need to) | 14:51 |
| pgavin | if it doesn't work with guile-2.0, then it's because the upstream doesn't support it | 14:52 |
| pgavin | so I wouldn't worry | 14:52 |
| stekern | yeah, look into = remove guile-2.0 and install guile-1.8 ;) | 14:53 |
| pgavin | ok :) | 14:53 |
| stekern | ok, gas seems to work fine | 14:54 |
| _franck_ | is there a way to invalidate the whole icache in one go (I think there was a topic on this point on the mailing list or forum, can't remember) ? | 15:00 |
| jonibo|laptop | _franck_: no | 15:02 |
| _franck_ | ok thanks :) | 15:02 |
| jonibo|laptop | if you're on the or1200 you can use a trick... | 15:02 |
| _franck_ | yes I am, so tell me please | 15:03 |
| jonibo|laptop | yup | 15:03 |
| jonibo|laptop | the or1200 implementation is a bit broken | 15:03 |
| jonibo|laptop | so it will invalidate a cacheline if just the low bits in the effective address match the line | 15:03 |
| jonibo|laptop | so it's enough to loop over the low addresses (in 16 byte increments) in order to clear the cache | 15:03 |
| _franck_ | low how ? | 15:04 |
| jonibo|laptop | it's a bit quicker than looping over the entire memory space | 15:04 |
| _franck_ | the 16 low bits ? | 15:05 |
| jonibo|laptop | yeah, only the low 15 bits match for 32kB cache | 15:05 |
| jonibo|laptop | it doesn't matter what high bits you have in the effective address you try to invalidate | 15:05 |
| _franck_ | ok.....thanks for the info | 15:06 |
| jonibo|laptop | i.e. invalidating 0x0 and 0x8000 clear the same cacheline, irregardless of what address is really cached there | 15:07 |
| jonibo|laptop | good luck | 15:07 |
| _franck_ | it would be good to have an invalidate all. Becaause when you download a progrm via gdb, then you download another one you should invalidate the whole thing | 15:07 |
| _franck_ | we don't do it in openocd (I didn't look at the or_debug_proxy) | 15:08 |
| jonibo|laptop | i agree... others don't.. it's been a topic of discussion on several occasions | 15:08 |
| _franck_ | and it is a waste of time to loop for invalidate | 15:08 |
| jonibo|laptop | especially if you do it correctly and do the entire 32-bit memory space | 15:09 |
| _franck_ | yes I think I've seen such discussion but couldn't fin dit | 15:09 |
| _franck_ | s/fin/find | 15:09 |
| pgavin | anyone happen to know why l.rfe has l.nops immediately after it (as if it has a delay slot) in the newlib sources? | 15:09 |
| jonibo|laptop | it's often unclear whether rfe has a delay slot or not | 15:09 |
| jonibo|laptop | i know it doesn't, but i've got l.nop after it here and there, too | 15:10 |
| pgavin | unclear in what way? | 15:10 |
| jonibo|laptop | hard to remember :) | 15:10 |
| pgavin | lol | 15:10 |
| pgavin | k | 15:10 |
| jonibo|laptop | it feels like a jump! | 15:10 |
| pgavin | lol, sorta | 15:10 |
| _franck_ | may be that can help: http://www.mail-archive.com/[email protected]/msg00515.html | 15:10 |
| _franck_ | arf you posted it | 15:11 |
| _franck_ | sorry :) | 15:11 |
| pgavin | well, I posted in that thread :) | 15:11 |
| _franck_ | yeah I've seen it after I clicked "paste" :D | 15:11 |
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!