Fallenou | 25 minutes to generate official Milkymist SoC bitstream using a debian virtualized in virtualbox on my mac book pro Core 2 Duo 2.4 GHz | 00:35 |
---|---|---|
Fallenou | lekernel_: 25 min to generate bitstream, does it sound a lot to you ? | 00:35 |
wpwrak | sounds like you have a fast machine :) | 00:36 |
wpwrak | it also takes ~25 min on my Q6600, native | 00:37 |
Fallenou | ok so i'm not in such a bad position then | 00:39 |
wpwrak | naw, you're suffering just like the rest of us :) | 00:40 |
Fallenou | I can't blame my computer if mmy design does not go as fast as I expect :) | 00:40 |
Fallenou | mmu* | 00:40 |
wpwrak | hehe :) | 00:40 |
Fallenou | I confirm timings are met using official milkymist git branch | 00:41 |
Fallenou | I must have done something wrong | 00:41 |
Fallenou | will have a look | 00:41 |
Fallenou | wpwrak: do you know if lekernel_ is using xst or synplify to generate nowadays Milkymist SoC bitstreams ? | 00:47 |
Fallenou | it seems Xst is the default ATM in the makefile | 00:47 |
wpwrak | i would think he uses the makefile, yes. but i don't know for sure | 00:48 |
Fallenou | ok | 00:48 |
Fallenou | I've heard about something like Xst generating errors in lm32 caches that got solved using synplify but don't know if it's still the case with recent ISE milestones | 00:49 |
Fallenou | OK latest Milkymist SoC seems to meet timings with ISE 13.3 according to http://www.milkymist.org/wiki/index.php?title=Recommended_ISE_versions | 00:50 |
Fallenou | I have 13.2 | 00:50 |
Fallenou | gn8 ! | 01:11 |
wolfspraul | n8 | 01:14 |
wolfspraul | I downloaded 13.4 recently, should give it a try... | 01:14 |
Fallenou | will try with it tomorrow, I need to backup 13.2 install (13 GB) before | 01:16 |
Fallenou | in case I have to go back to it | 01:16 |
Fallenou | don't have enough space to have both installed on my virtual machine :) | 01:16 |
wolfspraul | I think lekernel_ is on 13.3 (that's what the wiki page says) | 01:23 |
wolfspraul | the latest now is 13.4 | 01:23 |
Action: kristianpaul too | 01:23 | |
wolfspraul | if you do that you may be the first one with 13.4 | 01:24 |
wolfspraul | jus tfyi | 01:24 |
roh | wolfspraul: do you have contact to the people doing this: http://www.elinux.org/Embedded_Open_Modular_Architecture ? | 01:44 |
kristianpaul | roh: the guys that will beat the 39usd pi compuiter? | 01:48 |
roh | yes | 01:48 |
kristianpaul | irc channerl is decert.. | 01:48 |
roh | and since it seems to be a sincere grassroots attemt i rather check out that than the pi | 01:49 |
kristianpaul | but the guy already wrote to mail list | 01:49 |
kristianpaul | you rememnber mail in #qi-hardware ML about A10 stuff and fped | 01:49 |
roh | r-pi means no specs, no schems, no datasheets, and only partial opensource drivers from what we know now | 01:49 |
kristianpaul | sure | 01:49 |
roh | kristianpaul: thats why i am asking.. if there is talking happening ;) | 01:50 |
kristianpaul | well, by talking the guy is very quiet nevr reply even the mail thread.. | 01:50 |
kristianpaul | guess to busy hacking its coming boards to reply mails :) | 01:50 |
kristianpaul | also i and considering it is milkymist channel, i will care more of having a working ethernet driver for uclinux milkymist port | 01:56 |
kristianpaul | and a _VERY_ rought starting point for a freedom box with that :) | 01:56 |
kristianpaul | freedom hacker box ;) | 01:57 |
roh | hrhr | 01:58 |
kristianpaul | /\/\-:) | 02:06 |
wpwrak | lekernel_: "regf" (in various comments in compiler.h) == "register file" ? | 05:35 |
stekern | wpwrak: (u-boot) why would you need locking in a program that is singel threaded and (hardly) doesn't use interrupts? | 08:20 |
stekern | I kind of agree with the swiss army argument though ;) | 08:21 |
stekern | I've devoted some time into the openrisc u-boot port, mostly out of PR reasons | 08:22 |
stekern | it's a well known project, and I believe in wolfspraul's comment about "making people feel safe" | 08:22 |
lekernel_ | wpwrak: yes | 09:28 |
Fallenou | good news, timings are met with 13.4 but I did not test to flash and boot the generated bitstream | 09:47 |
Fallenou | +yet | 09:47 |
Fallenou | damn, git head of urjtag does not seemt to compile (under debian squeeze) | 11:23 |
Fallenou | misses libftd2xx.h | 11:25 |
wolfspraul | I just recently built it, let me see what I installed | 11:27 |
Fallenou | http://forum.sparkfun.com/viewtopic.php?p=50378 | 11:27 |
Fallenou | will try this | 11:28 |
wolfspraul | apt-get install dh-autoreconf autoconf libtool pkg-config bison flex python2.6-dev libftdi-dev libusb-1.0-0-dev | 11:29 |
wolfspraul | also apt-get build-dep urjtag | 11:29 |
wolfspraul | try those | 11:29 |
Fallenou | ok thanks | 11:30 |
wolfspraul | run ./autogen.sh --with-libusb --with-libftdi | 11:34 |
Fallenou | wo it seems ok now | 11:35 |
Fallenou | thanks | 11:35 |
Fallenou | oh no, same problem | 11:36 |
larsc | config.log says libftdi was found? | 11:37 |
Fallenou | yes libraries are found (libusb yes, libftdi yes (no async mode), libftd2xx yes) | 11:39 |
Fallenou | but the header is missing anyway | 11:39 |
Fallenou | will copy it manually from ftdi website | 11:40 |
wolfspraul | you can search it with apt-file search libftd2xx.h | 11:40 |
wolfspraul | btw I think you don't need this 2xx stuff | 11:40 |
wolfspraul | apt-file cannot find libftd2xx.h | 11:42 |
wolfspraul | I'm running my ./autoconf.sh again, and libftd2xx is :no | 11:42 |
Fallenou | even with the header from official website it fails | 11:44 |
Fallenou | oh ok I will disable it then | 11:44 |
stekern | you need to build explicitly without the ftd2xx stuff | 11:44 |
Fallenou | yep I'm trying | 11:45 |
larsc | wolfspraul: libftd2xx is the thing from the FTDI themself | 11:45 |
stekern | I had problems to when those got pulled in | 11:45 |
Fallenou | unrecognized option "--without-libftd2xx" | 11:45 |
Fallenou | damn | 11:45 |
Fallenou | ok just --without-ftd2xx | 11:47 |
Action: Fallenou needs to sleep more | 11:47 | |
Fallenou | will update the wiki to add --without-ftd2xx | 11:48 |
Fallenou | but after lunch | 11:49 |
Fallenou | bbl | 11:49 |
wpwrak | stekern: it tends to gets multithreaded and interrupt-driven as soon as you add USB ;-) | 12:23 |
stekern | wpwrak: oh, you are speaking about the usb-implementation... I feel your pain ;) | 12:28 |
stekern | but as long as your driver doesn't do interrupts, it should still be pretty much just polling | 12:30 |
wpwrak | yeah, as long as you don't use u-boot for much, it's pretty okay. like the plastic swiss army knife. as long as you don't open it, it'll be in good shape forever ;-) | 12:35 |
stekern | you shouldn't be backtalking the swiss army knives though, my 20 year old trusty victorinox still has its corkscrew in excellent shape! (and it has been opened on several occassions) | 12:41 |
wpwrak | i'm talking more of the malaysia-made ones ;-) | 13:02 |
stekern | heh, ok | 13:04 |
stekern | anyways, in my openrisc-m1soc merging adventure I've taken the path of porting BIOS to openrisc rather than porting the m1soc to u-boot | 13:09 |
wpwrak | good :) | 13:10 |
stekern | I deemed that to be easier, and I'm lazy ;) | 13:11 |
wpwrak | laziness, the engineer's #1 virtue | 13:11 |
stekern | I agree | 13:12 |
Fallenou | do I need svf or bsdl support in my urjtag ? | 15:38 |
Fallenou | cause none of both compiles | 15:38 |
kristianpaul | svf is not just for cpld? | 16:00 |
Fallenou | ok got urjtag to work and flashed the bitstream successfully | 16:15 |
kristianpaul | good | 16:17 |
wolfspraul | Fallenou: cool! it seems you are making massive progress! | 16:35 |
Fallenou | yep ! | 16:37 |
Fallenou | I'm getting good to go :) | 16:37 |
Action: Fallenou heating up | 16:37 | |
Fallenou | wiki updated | 17:02 |
Fallenou | ftdi chip is hotter than fpga | 17:31 |
lekernel | Fallenou: you shouldn't need to flash your bitstream, you can load it directly with pld load xxx.bit | 17:47 |
Fallenou | by flashing I meant loading it | 17:48 |
Fallenou | I just did 'make load-bitstream' | 17:48 |
Fallenou | mmu branch meets timing with ISE 13.4 :o good | 17:48 |
Fallenou | now need to write FSM for TLB management (check/miss/update/flush) | 17:49 |
lekernel | ok, just make sure that all hits take only 1 cycle | 17:50 |
Fallenou | ok | 17:51 |
Fallenou | I hope tlb miss won't happen too often | 17:52 |
Fallenou | because it would generate exception, kernel would have to lookup pfn using memory loads and then update tlb line and then return | 17:54 |
Fallenou | :/ | 17:54 |
Fallenou | lekernel: would an update to a ITLB line update the DTLB line as well ? | 18:16 |
Fallenou | or would we just want to be able to update independently each TLB | 18:16 |
Fallenou | I think we could gain having two different contents in those two TLBs | 18:17 |
Fallenou | so updating ITLB and DTLB independently | 18:17 |
lekernel | both I and D caches and TLBs operate independently | 18:17 |
Fallenou | ok | 18:17 |
Fallenou | just wanted to be sure :) | 18:17 |
kristianpaul | ah m1nor need to rewritten | 18:22 |
kristianpaul | to m1load | 18:22 |
kristianpaul | lekernel: sorry for mistype words but can you please stop writing private mails everytime you get angry by reading others mistakes! | 18:35 |
kristianpaul | sorry OT | 18:36 |
Fallenou | well, maybe it's better keeping it private rather than sending corrections each time on the public list, isn't it ? | 18:38 |
kristianpaul | sorry | 18:40 |
kristianpaul | sorry all | 18:41 |
Action: kristianpaul checks how to use spell checker inside mutt | 18:45 | |
kristianpaul | s/mutt/vim | 18:46 |
Fallenou | yep spell check is helpfull sometime :) | 18:46 |
Fallenou | -l ;-) | 18:46 |
larsc | +s | 18:49 |
larsc | ;) | 18:49 |
Action: larsc needs a spell checker for irssi | 18:49 | |
kristianpaul | larsc: ha mee too ;) | 18:50 |
Fallenou | It will be fun to test if MMU is working properly | 19:27 |
Fallenou | either modifying RTEMS, or Linux .... or a custom bios | 19:28 |
Fallenou | maybe a custom bios | 19:28 |
Fallenou | is it possible to reboot into bios ? | 19:29 |
Fallenou | holding a button or something | 19:29 |
Fallenou | instead of booting to rtems shell/FNoise | 19:29 |
lekernel | yes, press esc | 19:29 |
lekernel | https://twitter.com/#!/dev_dsp/status/161695220843560960/photo/1 | 19:30 |
Fallenou | oh, I need a keyboard : | 19:32 |
Fallenou | p | 19:32 |
larsc | press all three buttons | 19:32 |
Fallenou | it reboots into rtems | 19:33 |
lekernel | it works with the serial console, too | 19:34 |
lekernel | wpwrak: by the way, how do you plan to implement "hard" reset if there's only one button? | 19:36 |
kristianpaul | time pressed? | 19:36 |
larsc | havn't you seen his statemachine for the button behaviour? | 19:37 |
kristianpaul | trying to find it.. | 19:37 |
Fallenou | if I boot pressing escape in the uart it does not boot | 19:37 |
kristianpaul | launch flterm first | 19:38 |
lekernel | Fallenou: which is what you want, no? | 19:38 |
kristianpaul | ah, you mean boot from serial? | 19:38 |
larsc | pressing enter does the same thing btw | 19:38 |
Fallenou | lekernel: I want to have the bios prompt | 19:38 |
Fallenou | it does not boot the bios either | 19:38 |
lekernel | wtf? | 19:38 |
Fallenou | it just does nothing | 19:38 |
lekernel | h | 19:39 |
lekernel | of course you should press ESC | 19:39 |
lekernel | *while the board is rebooting* | 19:39 |
wpwrak | lekernel: something along these lines: http://downloads.qi-hardware.com/people/werner/tmp/onebutton-fsm.pdf | 19:39 |
lekernel | i see | 19:41 |
lekernel | great! | 19:41 |
lekernel | and I guess we map that to the middle button of the existing boards? and keep the current behaviour for the other two buttons? | 19:42 |
Fallenou | hum strange, I tried unplugging power supply, keeping ESC pressed (in uart, not a real USB keyboard plugged to M1), then plug power supply again, it prints a few letters and then nothing more | 19:42 |
wpwrak | sounds good to me | 19:42 |
lekernel | Fallenou: you can use reset instead of unplugging the power supply, and what are those letters? | 19:43 |
lekernel | you can also use Q instead of ESC | 19:43 |
kristianpaul | reset or make boot :) | 19:43 |
lekernel | (capital Q) | 19:43 |
wpwrak | larsc: btw, thanks for reminding me about my FSM - i had already forgotten about it ;-)) | 19:45 |
Fallenou | http://pastebin.com/6FhJ8Y3U | 19:46 |
wpwrak | Fallenou: the Esc is not very timing-critical. it's sufficient to press if when you see the first sign of life from the BIOS | 19:46 |
Fallenou | this is with holding capital Q when booting | 19:46 |
wpwrak | pressing something while booting may uncover unexplored dangers :) | 19:46 |
wpwrak | hmm, just one press should do | 19:47 |
Fallenou | oh damn | 19:47 |
Fallenou | it's just minicom crap | 19:47 |
lekernel | Fallenou: use flterm | 19:47 |
Fallenou | using flterm I get the bios | 19:47 |
Fallenou | yes | 19:47 |
Fallenou | good | 19:47 |
lekernel | yes, that's also why I wrote flterm... iron out all this serial port mess | 19:47 |
Fallenou | I think the bios will be helpful to debug mmu | 19:48 |
wpwrak | ah, you wrote flterm. interesting. | 19:48 |
Fallenou | simple program, easy to check program control flow | 19:48 |
lekernel | or you can find out what set of minicom options happens to work, if you have some time to waste | 19:48 |
wpwrak | how about making it easier to pick the gdb pass-through device ? not sure what would be the best approach. perhaps setting a symlink specified by the user or such | 19:49 |
Fallenou | ooooh I get all the bios prelude now when booting ! flterm rocks | 19:49 |
kristianpaul | Fallenou: ;) | 19:52 |
kristianpaul | flterm plus jtag-boot | 19:52 |
Fallenou | jtag-boot ? what is it ? | 19:56 |
Fallenou | serialboot ? | 19:56 |
kristianpaul | nope | 19:58 |
kristianpaul | Fallenou: http://projects.qi-hardware.com/index.php/p/wernermisc/source/tree/master/m1/jtag-boot | 19:59 |
kristianpaul | make boot; make standby; life esier ! | 19:59 |
Fallenou | what does it do exactly ? | 20:01 |
kristianpaul | avoid you push buttons | 20:01 |
Fallenou | well make load-bistream loads a new bitstream and reboots the system without pushing any button | 20:02 |
wpwrak | yeah, jtag-boot is for when you have the bitstream in NOR. also, if you have the new standby bitstream that doesn't wait for a button, you can just do a pld reconfigure | 20:14 |
Fallenou | hum ok | 20:14 |
Action: Fallenou just understood that he needs to modify lm32-gnu-as to add support for MMU CSR | 20:15 | |
wpwrak | (pld reconfigure) instead of jtag-boot. not instead of pld load. | 20:16 |
Fallenou | are you using gcc 4.5.2 or 4.5.3 ? | 20:18 |
kristianpaul | may be too early to ask, but this MMU will be optional in the future? :) | 20:20 |
Fallenou | imo it will easily be optional | 20:37 |
Fallenou | just like ICACHE and DCACHE | 20:37 |
larsc | lekernel: any better idea how to find out which actors may qualify for timesharing than comparing each actor which every other actor? | 21:15 |
larsc | finding a valid schedule fast is the next problem | 21:17 |
kristianpaul | lekernel: had you ever used TIG in constraint file? ;) | 21:32 |
Action: kristianpaul wonder if lattice have such extensive documentation as xilinx does | 21:35 | |
lekernel | larsc: no, not really... | 21:38 |
lekernel | kristianpaul: yes, see tdc core | 21:38 |
kristianpaul | ah intereting | 21:40 |
--- Mon Jan 30 2012 | 00:00 |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!