Topic for #milkymist is now Radical Tech Coalition :: Milkymist One, Migen, Milkymist SoC & Flickernoise :: Logs: http://en.qi-hardware.com/mmlogs
cladamw has joined #milkymist
xiangfu has joined #milkymist
rejon has joined #milkymist
DJTachyon has joined #milkymist
<GitHub124> [flickernoise] wpwrak pushed 4 new commits to imaginarium: http://git.io/M1ql4Q
<GitHub124> [flickernoise/imaginarium] images: added animation demo in patches/demo/wheel/ - Werner Almesberger
<GitHub124> [flickernoise/imaginarium] images: updated regression tests for removal of two image file limit - Werner Almesberger
<GitHub124> [flickernoise/imaginarium] compiler: use named regular expression for identifiers - Werner Almesberger
DJTachyon has joined #milkymist
pablojavier has joined #milkymist
pablojavier has quit [#milkymist]
bkero has joined #milkymist
cladamw has joined #milkymist
wolfspraul has joined #milkymist
fpgaminer has joined #milkymist
rejon has joined #milkymist
Martoni has joined #milkymist
* lekernel_ just found another map bug
rejon has joined #milkymist
elldekaa has joined #milkymist
wolfspraul has joined #milkymist
Jia has joined #milkymist
<Jia> hi all is there unfixed gcc bugs?
<lekernel> hi Jia, welcome
<Jia> lekernel: hi
<lekernel> Jia: yes... just try to build the SVN head for LM32
<lekernel> we also have a little list on http://www.milkymist.org/wiki/index.php?title=GCC_bug_list
<xiangfu> Jia, lekernel is Sebastien :)
<xiangfu> Jia, Welcome.
<lekernel> also, C++ has various problems which kick in when compiling complex code; for example, when attempting to compile Qt4 the wrong object methods get called at times
<lekernel> Jia: about your LLVM patch I'll wait a bit for JP to have a look at it
<Jia> lekernel: there is something wrong.
<lekernel> yes, we received your second message too
<Jia> -*- C++ -*- should NOT be cpp files.
<Jia> maybe I can help.
<xiangfu> Jia, Hi
wolfspraul has joined #milkymist
<xiangfu> Jia, about test: this bug : http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43807 lm32-elf infinite loop will let you can't create correct result.
<Jia> sorry for my delay reply.
<wolfspraul> Jia: hey Jia, nice to see and meet you! :-)
<Jia> I have to go.
<Jia> wolfspraul: nice to meet you too!
<wolfspraul> no worries, we are survivors. you can leave us in the desert, come back 3 months later and we are still alive :-)
<Jia> and I will try to fix the bugs.
<wolfspraul> that would be fantastic, any little help actually
<wolfspraul> let's take it slowly
<wolfspraul> do you have a Milkymist One or are interested in having one?
<Jia> I just working on gcc&llvm.
<wolfspraul> how do you test stuff right now?
<wolfspraul> yes sure, that's great!
<Jia> maybe xiangfu will help me.
<xiangfu> Jia, on test, sure.
<xiangfu> ( infinite loop ) this bug is every easy test. :)
<lekernel> Jia: by the way, we also have a good QEMU port
<Jia> sorry all. I have to go now. see you later.
<wolfspraul> sure
<lekernel> which can be useful to test gcc/llvm
<wolfspraul> stop by here more often...
<wolfspraul> (in this channel)
<GitHub114> [milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/nFdVpA
<GitHub114> [milkymist-ng/master] s6ddrphy: use single-ended DQS - Sebastien Bourdeauducq
<GitHub29> [milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/WEnjnw
<GitHub29> [milkymist-ng/master] Connect DDR PHY - Sebastien Bourdeauducq
<GitHub196> [migen] sbourdeauducq pushed 1 new commit to master: https://github.com/milkymist/migen/commit/a1ad30faab428905349214643837a1deed861b20
<GitHub196> [migen/master] fhdl/verilog: properly connect instance inouts - Sebastien Bourdeauducq
wolfspraul has joined #milkymist
VJTachyon has joined #milkymist
wolfspraul has joined #milkymist
wolfspraul has joined #milkymist
kilae has joined #milkymist
antgreen has joined #milkymist
wolfspraul has joined #milkymist
<wpwrak> ftp> put ../../pleiades.jpg -> death. loops in cpukit/libfs/src/imfs/imfs_eval.c:IMFS_eval_path
<wpwrak> and they're putting this stuff on space missions ? :)
<GitHub18> [flickernoise] wpwrak pushed 8 new commits to imaginarium: http://git.io/4RvCCA
<GitHub18> [flickernoise/imaginarium] test/imgtag: tests for the image tag syntax - Werner Almesberger
<GitHub18> [flickernoise/imaginarium] ptest.c: added section comments - Werner Almesberger
<GitHub18> [flickernoise/imaginarium] ptest: in parse-only mode, print value of constants - Werner Almesberger
<GitHub147> [milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/TL1UUQ
<GitHub147> [milkymist-ng/master] Map DDR PHY controls in CSR - Sebastien Bourdeauducq
<lekernel> wpwrak: yeah, the filesystem stuff is amazingly crappy
<lekernel> I can't think of anything to defend it
<lekernel> lots of code duplication, lots of bugs, major design flaws (like lack of synchronization), lousy API, ...
<wpwrak> heh, so there may be more fun waiting for us :)
<lekernel> phew, don't bother
<lekernel> unless you want to redesign the RTEMS filesystem from scratch :)
<lekernel> seriously there's nothing good in it
<lekernel> better spend that time on linux
<lekernel> the current solution.. somehow works
<wpwrak> well, that hang is kinda nasty. you can produce it by accident. let's see if it's easy to fix ...
<lekernel> even their main developers admit it's crappy and most of it is in that state due to historical accidents
<lekernel> but the redesign exceeds the available motivation
<wpwrak> that's sort of impressive in itself :)
<wpwrak> an operating system without a working file system. erm ;-)
<Hodapp> DOS was arguably that.
<Hodapp> ...so was whatever the OS was that Apple Newtons ran
<lekernel> so, the canonical place for FJMEM is ${HOME}/.qi/milkymist/fjmem/fjmem.bit it seems?
<wpwrak> is it ?
<wpwrak> i have it in $HOME/.qi/milkymist/release/2011-07-13/fjmem.bit and $HOME/.qi/milkymist/release/current/fjmem.bit
* kristianpaul stick with bare-metal apps for now
<wpwrak> and it probably should be in /usr{/local,}/share/something
<lekernel> /usr/local/share/milkymist/fjmem.bit ?
<wpwrak> that sounds much cleaner to me, yes.
<lekernel> can you add it to m1nor?
<lekernel> maybe also with the /usr/share - if someday distros want to package fjmem by some miracle
<GitHub63> [milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/aEU4EQ
<GitHub63> [milkymist-ng/master] Add build Makefile and JTAG load script - Sebastien Bourdeauducq
<GitHub99> [milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/R55_Yg
<GitHub99> [milkymist-ng/master] bios: add flash target using m1nor - Sebastien Bourdeauducq
<wpwrak> updated and tested. at least for now, i search for the $HOME/.qi/ paths first
<lekernel> thanks!
<wpwrak> thanks for giving the file an official home :)
<GitHub31> [milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/1FI7-A
<GitHub31> [milkymist-ng/master] bios: DDR initialization skeleton - Sebastien Bourdeauducq
azonenberg has joined #milkymist
lekernel_ has joined #milkymist
antgreen has joined #milkymist
<GitHub41> [milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/qKBPcQ
<GitHub41> [milkymist-ng/master] DFI injector (untested) - Sebastien Bourdeauducq
<GitHub199> [migen] sbourdeauducq pushed 3 new commits to master: https://github.com/milkymist/migen/compare/a1ad30f...55a265d
<GitHub199> [migen/master] fhdl: check we pass BV to signals - Sebastien Bourdeauducq
<GitHub199> [migen/master] bus: add interconnect statements function - Sebastien Bourdeauducq
<GitHub199> [migen/master] bank: add RE signal for registers made of fields - Sebastien Bourdeauducq