lekernel changed the topic of #m-labs to: Mixxeo, Migen, MiSoC & other M-Labs projects :: fka #milkymist :: Logs http://irclog.whitequark.org/m-labs
<GitHub103> [migen] sbourdeauducq pushed 1 new commit to master: http://git.io/rsVxcw
<GitHub103> migen/master 09ebcc4 Florent Kermarrec: setup.py: fix README filename
fengling has joined #m-labs
<rjo> sb0: yes it's gated by the decoded address.
rjo has quit [Quit: leaving]
rjo has joined #m-labs
fengling has quit [Ping timeout: 268 seconds]
fengling has joined #m-labs
ccier has joined #m-labs
ccier has quit [Client Quit]
<rjo> mithro: are you looking into migen/misoc for hdmi2usb?
early has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
early has joined #m-labs
<GitHub29> [ARTIQ] sbourdeauducq pushed 5 new commits to master: http://git.io/t-I88Q
<GitHub29> ARTIQ/master 16b2d9f Sebastien Bourdeauducq: runtime: support real-time FUD
<GitHub29> ARTIQ/master 813bc90 Sebastien Bourdeauducq: rtio: support readout of counter from software
<GitHub29> ARTIQ/master b5f145d Sebastien Bourdeauducq: devices: new RTIO driver
<GitHub75> [ARTIQ] sbourdeauducq pushed 1 new commit to master: http://git.io/dF4utA
<GitHub75> ARTIQ/master 6861d28 Sebastien Bourdeauducq: runtime: support arbitrarily long initial DDS setup time
<sb0> rjo, how should we deal with the fact that the FUD resolution (8ns) is lower than the RTIO resolution (1ns)? silently round time to the nearest FUD step?
<mithro> rjo: yes
_florent_ has joined #m-labs
<GitHub175> [misoc] sbourdeauducq pushed 1 new commit to master: http://git.io/r8F04Q
<GitHub175> misoc/master 503a2f0 Sebastien Bourdeauducq: mor1kx: sync
<GitHub105> [ARTIQ] sbourdeauducq pushed 1 new commit to master: http://git.io/Kt_G7w
<GitHub105> ARTIQ/master b207a3c Sebastien Bourdeauducq: rtio: remove ISE bug workaround
uf6667 has joined #m-labs
<sb0> hi uf6667
<_florent_> hi
<_florent_> stekern: here is a small patch for the store buffer that seems to give interesting result (at least on misoc-de0nano)
<uf6667> hi sb0
<_florent_> I was not able to test it by it should have the same behaviour
<_florent_> and the report for the last commits on misoc-de0nano
bhamilton has joined #m-labs
bhamilton is now known as Guest30013
<sb0> uf6667, I'm curious what about you are using migen for :)
<uf6667> lol why? :)
<uf6667> I'm trying to catch up with openvizsla
Guest30013 has quit [Remote host closed the connection]
xiangfu has quit [Ping timeout: 260 seconds]
<stekern> ?
<_florent_> yes, the results are in the previous link http://pastebin.com/HXxGFN6i
<stekern> ah, I see now
<stekern> you didn't happen to do a build on 2440fab too?
<stekern> err, 0454b5b I mean
<_florent_> started :)
<stekern> since when I built for the ppro target, I got an increase by ~400 LUTs by that(!)
<stekern> when by all logic, it should use less resources
<_florent_> yes I've tried that on mixxeo, and have the same result
<_florent_> it seems ISE is replicating lots of things...
<_florent_> the results with 0454b5b : http://pastebin.com/F6W5puHR
<stekern> so it's actually mariginally bigger there too..
<_florent_> yes strange...
<_florent_> let me check what it gives me if I revert the 2440 and use my last patch...
<stekern> maybe the tools were already able to re-use the registered outputs in the RAM
<stekern> but on a source level, explicitly using the read_enable logic is a lot cleaner, so I'm inclined to keep it as is now
uf6667 has quit [Ping timeout: 268 seconds]
<_florent_> yes I agree
<_florent_> hmm interesting:
<_florent_> if I use upstream version + my patch
<_florent_> I have better result than
<_florent_> upstream version - 2240 + my patch
<_florent_> so 2240 + my patch seems to really improve things finally
fengling has quit [Quit: WeeChat 1.0]
<stekern> the difference on ppro is -2 LUT ;)
<_florent_> lol, with my patch?
<stekern> yes
<stekern> I'll apply it anyway, since it seems to make a real difference on cyclone iv and the logic is as easy (or even slightly easier) to follow
<stekern> the only thing where it might harm is if there's a cost to reset to !0
<sb0> on s6, if the reset value is different than the bitstream initialization value, the reset input cannot be used and it needs to go to LUT logic
<sb0> I'm not sure what happens when you don't specify a bitstream init value - if it sets it to the reset value or not
<sb0> had you used migen, you could easily fix that problem by automatically inserting bitstream init values that match the reset values :-)
<stekern> I would assume it'd be set to the reset value
<stekern> iirc cyclone devices use a different approach, where they have a reset to 1 and a reset to 0 input
<stekern> ...but don't take quote me on that, I might very well be remembering wrong ;)
<stekern> s/take quote/quote
uf6667 has joined #m-labs
MY123 has joined #m-labs
<uf6667> sb0: are you looking for someone to make cores?
<uf6667> I specialize in timing constraint optimization... if that's of any use to you :)
<sb0> ok... what have you worked on before?
<sb0> timing constraint optimization is rather general :-)
<uf6667> hmmm plenty of stuff but the most recent was progskeet
<uf6667> made the usb core from scratch
<sb0> is the actel fpga connected directly to the usb port?
<uf6667> yes
<sb0> it's 12Mbps right?
<uf6667> yeah lmao
<uf6667> ok bad example :D
<sb0> so you did the serial interface engine, etc.
<uf6667> yes
<sb0> did you like it? :-)
<uf6667> pretty much, but it was kinda hard without analysis hardware xD
<sb0> cool. so you like disgusting protocol. that's a useful perversion to have. everytime I touch USB, I want to behead its designers.
<uf6667> had to embed a serial port into pretty much everything
<uf6667> haha :)
<uf6667> oh and everything related to stuff depending on counters
<sb0> do you have the code of that USB core somewhere btw?
<uf6667> also made a 2chan oscilloscope using a s3e board and vga monitor, including triggering and all, without any software
<sb0> the download link on your website says sql error
<uf6667> yeah sure let me check
<uf6667> oh lol, its not opensource :)
<uf6667> and its not my site anymore
<uf6667> I got ripped off hard
<uf6667> ah shit, I forgot my external hdd in england :/
<sb0> where do you live, btw?
<uf6667> luxembourg and studying in uk, you?
<sb0> (external hdd) that's one reason to make stuff opensource ;)
<sb0> hk
<uf6667> ah cool, useful :D
<sb0> I haven't had time to check out Shenzhen yet...
<uf6667> doesn't matter! If I lived there, I'd be making stuff 24/7 and throw everything into production
<uf6667> plus, studying chemistry, plenty of chemicals you can get there you can't get here :)
<sb0> what kind of thing do you want to put into production?
<uf6667> bugs ;)
<sb0> haha, my best country so far to order chemicals has been France
<sb0> and Germany the worst
<sb0> bugs?
<uf6667> really? :o
<uf6667> I need some LiAlH4 :D
<uf6667> bugs like spybugs
<uf6667> there's plenty of private detectives here
<uf6667> they go crazy for it
<uf6667> "need" meaning "I'd really really like some"
<sb0> hmm what do you want to do with LiAlH4?
<sb0> ah, well, that was funny when I was in high school...
<uf6667> yeah it's very profitable now :)
sh[4]rm4 has joined #m-labs
<sb0> not all profitable things are good...
<uf6667> true, but this is :D
sh4rm4 has quit [Ping timeout: 260 seconds]
<rjo> sb0: fud resolution. you mean the cast from artiq language set_frequency() to fud rtio time? yes. treat it like the rounding of rtio timings.
sh[4]rm4 is now known as sh4rm4
sb0 has quit [Ping timeout: 264 seconds]
sb0 has joined #m-labs
MY123 has quit [Quit: Connection closed for inactivity]
sh4rm4 has quit [Ping timeout: 260 seconds]
sh4rm4 has joined #m-labs
<_florent_> stekern, just for info:
<_florent_> this modification allows to store data in registers when DEPTH_WIDTH=1 on store buffer
<_florent_> this kind of remove the store buffer and seems to save 200 regs on mixxeo BaseSoC.
<_florent_> this one changes some logic in logic operations and seems to save 100 luts on mixxeo BaseSoC.
<_florent_> It's not tested nor simulated, but I'll try to do that soon, but in case you want to have a look...
_florent_ has quit [Quit: Page closed]
uf6667 has quit [Ping timeout: 255 seconds]
uf6667 has joined #m-labs
MY123 has joined #m-labs
uf6667 has quit [Ping timeout: 246 seconds]
mumptai has joined #m-labs
<GitHub119> [NetBSD] fallen pushed 3 new commits to master: http://git.io/oO15fg
<GitHub119> NetBSD/master b7d17d1 Yann Sionneau: uart: add TTY layer support
<GitHub119> NetBSD/master 1e167fe Yann Sionneau: clock driver: Fix softc handling. Hello world from user space works :)
<GitHub119> NetBSD/master 712cccd Yann Sionneau: init: update /sbin/init which is now able to print to uart console
Alain has joined #m-labs
MY123 has quit [Quit: Connection closed for inactivity]
uf6667 has joined #m-labs
Alain has quit [Quit: ChatZilla 0.9.90.1 [Firefox 31.0/20140716183446]]
nengel has quit [Ping timeout: 268 seconds]
mumptai has quit [Ping timeout: 268 seconds]
mumptai has joined #m-labs
mumptai has quit [Read error: Connection reset by peer]
nengel has joined #m-labs
mumptai has joined #m-labs
kyak_ has joined #m-labs
kyak_ has quit [Changing host]
kyak_ has joined #m-labs
kyak has quit [Ping timeout: 276 seconds]
juliusb_ has quit [Ping timeout: 276 seconds]
juliusb has joined #m-labs
sb0_ has joined #m-labs
nengel has quit [*.net *.split]
uf6667 has quit [*.net *.split]
sb0 has quit [*.net *.split]
aeris has quit [Ping timeout: 260 seconds]
nengel has joined #m-labs
sh4rm4 has quit [*.net *.split]
bentley` has quit [Ping timeout: 255 seconds]
clever has quit [*.net *.split]
ohama has quit [*.net *.split]
clever has joined #m-labs
aeris has joined #m-labs
nengel has quit [Ping timeout: 246 seconds]
wpwrak has quit [Ping timeout: 246 seconds]
Hawk777 has quit [Ping timeout: 246 seconds]
ohama has joined #m-labs
gric_ has quit [Ping timeout: 276 seconds]
sb0_ has quit [Ping timeout: 276 seconds]
[florian] has quit [Ping timeout: 276 seconds]
wpwrak has joined #m-labs
nengel has joined #m-labs
sh4rm4 has joined #m-labs
sb0 has joined #m-labs
Hawk777 has joined #m-labs
rjo has quit [Ping timeout: 252 seconds]
rjo has joined #m-labs
sh4rm4 has quit [Ping timeout: 260 seconds]
mumptai has quit [Ping timeout: 260 seconds]
gric_ has joined #m-labs
sh4rm4 has joined #m-labs
Hawk777 is now known as Guest40586
17SAA6DVI has joined #m-labs
17SAA6DVI has quit [Ping timeout: 272 seconds]
mumptai has joined #m-labs
[florian] has joined #m-labs