| 00:04.59 | CP|too | gah - alsa uses hg |
| 00:05.00 | CP|too | I hate hg |
| 00:05.13 | *** join/#edev dijenerate (n=dijenera@72.22.150.235) |
| 00:06.59 | CP|too | oh - now mecurial has cloned the git web interface |
| 00:07.02 | CP|too | how original fellas |
| 00:15.44 | *** join/#edev joe_bleau (n=joe_blea@adsl-208-191-52-129.dsl.fyvlar.swbell.net) |
| 01:12.14 | MonMotha | CP|too: the huge assortment of VCSes is getting annoying |
| 01:51.20 | *** join/#edev dijenerate (n=dijenera@72.22.150.235) |
| 02:36.02 | *** join/#edev redblue (i=star@ppp210.108-253-207.mtl.mt.videotron.ca) |
| 02:55.11 | *** join/#edev Off_Namuh (i=GPS@gateway/tor/x-a75e351a9c3e147d) |
| 03:16.54 | *** join/#edev roxfan (n=dunno@63.198-201-80.adsl-dyn.isp.belgacom.be) |
| 04:08.43 | *** join/#edev mahr_ (n=mahr@208.100.151.210) |
| 04:43.48 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 04:53.01 | *** join/#edev marcj (n=marcj@c-67-176-34-134.hsd1.co.comcast.net) |
| 04:53.32 | *** part/#edev marcj (n=marcj@c-67-176-34-134.hsd1.co.comcast.net) |
| 06:17.25 | *** join/#edev ccjoe (n=ccjoe@c-67-166-124-229.hsd1.co.comcast.net) |
| 06:34.19 | *** join/#edev agaffney (n=agaffney@gentoo/developer/pdpc.active.agaffney) |
| 08:28.49 | *** join/#edev redblue (i=star@ppp114.108-253-207.mtl.mt.videotron.ca) |
| 08:32.06 | *** join/#edev linac (n=lin@59.61.126.18) |
| 08:37.49 | *** join/#edev hw (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 09:47.22 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 09:50.38 | *** join/#edev dijenerate (n=dijenera@72.22.150.235) |
| 10:48.02 | *** join/#edev SeJo (n=SeJo@freenode/staff/sejo) |
| 10:48.06 | *** part/#edev SeJo (n=SeJo@freenode/staff/sejo) |
| 11:45.05 | *** join/#edev dijenerate (n=dijenera@72.51.69.65) |
| 12:19.42 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 12:20.52 | T0mW | You ever get the impression that the person at the other end of the email is someone who is running the messages through Babblefish? |
| 12:28.27 | Crofton|home | emailing sub-continent outsourcers again? |
| 12:30.00 | T0mW | Crofton: chinese |
| 12:30.28 | Crofton|home | ah |
| 12:30.49 | T0mW | I ask for a cable to plug into their jtag connector, they send me a drawing of the jtag connector. |
| 12:31.12 | Crofton|home | :) |
| 12:31.17 | Crofton|home | but they were very polite |
| 12:31.51 | T0mW | Oh! You know that oddly formatted way that they sent you message content? The one that is tab-stopped? heh. |
| 12:32.10 | T0mW | just seems a little unreal |
| 12:33.36 | T0mW | LOL, I'm trying to think of the poor fellow at the other end of the "wire". heh, he has to deal with an American English message. And, I wonder how much of the message content is getting across the translation device / barrier/ |
| 12:33.48 | T0mW | oh well |
| 12:34.58 | T0mW | I know that I've seen those stupid cables before and it just bugs me that I don't recall where. |
| 13:08.02 | *** part/#edev TomW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 13:10.53 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 13:16.26 | *** join/#edev drmikecrowe (n=drmikecr@208.104.23.43) |
| 13:36.11 | *** join/#edev sunrunner20 (n=jamesdoe@pool-71-123-194-235.dllstx.fios.verizon.net) |
| 13:36.52 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 13:36.52 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:14.38 | *** join/#edev dijenerate (n=dijenera@72.51.69.65) |
| 14:21.17 | *** join/#edev chouimat|work (n=mchouina@firewall.oeone.com) |
| 14:54.14 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 14:58.47 | *** part/#edev Crofton|home (n=balister@66.207.66.26) |
| 14:58.53 | *** join/#edev Crofton|home (n=balister@66.207.66.26) |
| 14:59.54 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 15:07.19 | *** join/#edev drmikecrowe_ (n=drmikecr@208.104.23.34) |
| 15:21.00 | *** join/#edev key2 (n=Ritual@193.33.36.20) |
| 15:21.17 | prpplague | key2: greetings |
| 15:22.27 | key2 | yop |
| 15:22.43 | key2 | someone knows an ARM9 based CPU/dev kit at > 500Mhz ? |
| 15:32.03 | davidc__ | key2: IIRC ARM9's don;t go that fast |
| 15:32.16 | davidc__ | key2: but there is an ARM something gunstix that runs at 600 |
| 15:32.44 | key2 | MMU less |
| 15:32.44 | key2 | .. |
| 15:33.24 | davidc__ | key2: you want an MMU less one, or that one is mmu less? |
| 15:33.43 | davidc__ | its an xscale, so it should have an mmu |
| 15:35.14 | key2 | why would they run uClinux then |
| 15:35.22 | key2 | i need to run a linux 2.6 and not a uClinux |
| 15:37.18 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 15:40.27 | davidc__ | key2: the info I've been able to find indicates its running full linux 2.6 |
| 15:40.35 | davidc__ | key2: but that info could be wrong |
| 15:41.48 | key2 | mmh |
| 15:41.50 | key2 | i just looked |
| 15:41.51 | key2 | yeah |
| 15:41.53 | key2 | thats true |
| 15:42.25 | davidc__ | anyhow, the xscale docs also indicate that they have MMUs builtin, and I recall them being capable of running full linux |
| 15:42.42 | MothaWork | only things I'm aware of is the former Intel PXA servers |
| 15:42.44 | MothaWork | *series |
| 15:42.49 | MothaWork | now owned by Marvell, so good luck |
| 15:43.06 | davidc__ | but just FYI - you won't find an ARM9 at > ~220mhz - the 5 stage pipeline doesn't go much faster than about that on any shipping chips I've seen |
| 15:43.06 | MothaWork | there's also the IXP series, which may go that fast, but is specifically targetted at networking apps |
| 15:43.13 | MothaWork | those are also not ARM9s |
| 15:43.16 | davidc__ | yeah, this is a PXA |
| 15:43.24 | davidc__ | [on the gumstix] |
| 15:44.01 | davidc__ | I've overclocked one chip rated at 180 to 220ish [atmel at91] and it worked, but I doubt you could push it much farther :P |
| 15:44.17 | MothaWork | atmel' |
| 15:44.20 | key2 | davidc__: but does it support the ARM9 codeset ? |
| 15:44.24 | MothaWork | atmel's new at91 parts go up to 200 anyway |
| 15:44.36 | key2 | davidc__: if I have a code compiled for ARM9, would it run on xscale >? |
| 15:44.37 | key2 | ? |
| 15:44.53 | davidc__ | key2 - depends. User land code or kernel code? |
| 15:44.58 | key2 | both |
| 15:45.11 | MothaWork | I believe the ISAs are compatible, but some of the onboard peripherals (MMU, etc.) are not |
| 15:45.19 | davidc__ | user land is probably, kernel is much less likely |
| 15:45.29 | MothaWork | those tend to be somewhat device specific anyway, though |
| 15:45.32 | davidc__ | because of the peripherals / mmu that Motha mentioned |
| 15:46.05 | davidc__ | MothaWork: they hit 200? Interesting. The at91sam9260 was originally rated at 200mhz, then was rerated at 180, they back to 200 now? |
| 15:46.23 | MothaWork | davidc__: I know the 9263 datasheet still says 200 |
| 15:46.34 | key2 | mmh |
| 15:46.42 | key2 | well the kernel could be recompiled it's not a big deal |
| 15:46.43 | key2 | ;) |
| 15:46.45 | davidc__ | MothaWork: hm.. I'll have to checkout the latest '60 datasheet |
| 15:46.48 | MothaWork | it's a very new part, though |
| 15:47.03 | davidc__ | key2: you'd be best off using the stock gumstix kernel, as it already works |
| 15:47.19 | davidc__ | key2: unless you need RT functionality or something |
| 15:47.36 | davidc__ | 'later all |
| 15:47.45 | davidc__ | [my last day at work, so I should try to be on time :P] |
| 16:17.04 | *** join/#edev ico2 (n=ico2@modem-1038.chimpanzee.dialup.pol.co.uk) |
| 16:26.51 | *** join/#edev Crofton|work (n=balister@ossie.wireless.vt.edu) |
| 16:36.54 | prpplague | davidc__: hehe |
| 16:38.41 | prpplague | key2: http://www.elinux.org/ARM_Processor |
| 16:38.57 | prpplague | key2: thats a charge i made with some data on the various pipe depths and functionality |
| 16:39.15 | prpplague | key2: the xscale line has a larger pipe depth than the arm9 |
| 16:39.25 | prpplague | key2: which allows it to run at fast clock speeds |
| 16:39.49 | *** join/#edev ico2 (n=ico2@modem-1038.chimpanzee.dialup.pol.co.uk) |
| 16:39.52 | prpplague | key2: problem is that due to the larger pipe, you create a problem with it needing more setup time to populate the pipe |
| 16:40.26 | prpplague | key2: so in essense, an xscale running at 600mhz is generally equal in performance to a 400mhz arm9 |
| 16:58.53 | MonMotha | prpplague: would depend on how branchy your code is and how well it predicts the branch |
| 16:59.09 | *** join/#edev hw__ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 17:03.46 | prpplague | MonMotha: that is true, but for example, doing most |
| 17:04.11 | prpplague | MonMotha: basic applications in both linux and wince, we get about the same performance between a s3c2440 and a px270 |
| 17:05.11 | MonMotha | prpplague: understood. That seems reasonable for the situation |
| 17:05.37 | MonMotha | though given that he is specifying a processor by clock speed, it may be that he can optimize the main loop to not trash the pipeline too badly |
| 17:06.11 | prpplague | MonMotha: indeed |
| 17:07.07 | MonMotha | and that should have been thrash, not trash :) |
| 17:11.50 | MonMotha | also, why the hell did Intel sell the PXA series to Marvell? |
| 17:14.45 | chouimat|work | hi MonMotha prpplague |
| 17:15.52 | MonMotha | hi |
| 17:15.53 | prpplague | chouimat|away: greetings |
| 17:15.56 | MonMotha | brb |
| 17:16.14 | prpplague | MonMotha: they were trying to get out of markets that they weren't good at |
| 17:22.53 | MothaWork | ok, at work |
| 17:22.59 | prpplague | MonMotha: hehe |
| 17:23.04 | MothaWork | prpplague: they were decent procs, and now that Marvell has them, we can't use them |
| 17:23.11 | prpplague | indeed |
| 17:23.27 | MothaWork | also has me worried about what might happen to IXP |
| 17:23.33 | prpplague | it was the best thing that could have happened for samsung |
| 17:25.38 | MothaWork | hum, what's a good 256k/512k ARM7TMDI with USB device, ethernet, and SPI which is available in a package more conducive to hand soldering than a 0.5mm QFP? |
| 17:25.46 | MothaWork | I think I can do a .5mm QFP, but it's tough |
| 17:25.53 | MothaWork | even .65mm QFPs are easier |
| 17:26.17 | prpplague | hmm, i rarely use something that small |
| 17:26.22 | MothaWork | price is not a major factor since I'm making like 3 |
| 17:26.27 | prpplague | MothaWork: i'd check out whats available from sparkfun |
| 17:26.53 | MothaWork | I'm thinking of the Atmel SAM7X series, but I was curious if there was something in a better package. NXP has some suitable parts, but they also seem to be in tiny packages |
| 17:30.07 | MothaWork | hum, our internet seems to be having issues |
| 17:30.19 | MothaWork | we're on a wireless link to a college across the street |
| 17:30.56 | MothaWork | I *think* I can solder a .5mm QFP |
| 17:31.03 | MothaWork | need to get the footprint right though |
| 17:49.05 | *** join/#edev chouimat|work (n=mchouina@firewall.oeone.com) |
| 19:17.30 | *** join/#edev dijenerate (n=dijenera@72.22.150.235) |
| 19:20.26 | MothaWork | hey, anybody perchance have any recommendations on how to gap ferrite transformer cores |
| 19:20.27 | MothaWork | ? |
| 19:54.46 | *** join/#edev drmikecrowe_ (n=drmikecr@208.104.23.34) |
| 20:28.28 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 20:32.17 | *** join/#edev chouimat (n=mchouina@firewall.oeone.com) |
| 21:06.24 | *** join/#edev Crofton (n=balister@hc652165a.dhcp.vt.edu) |
| 21:08.06 | *** join/#edev agaffney (n=agaffney@gentoo/developer/pdpc.active.agaffney) |
| 21:09.51 | *** join/#edev agaffney_ (n=agaffney@68-188-61-127.dhcp.stls.mo.charter.com) |
| 22:45.42 | *** join/#edev Crofton (n=balister@66.207.66.26) |
| 23:34.52 | *** join/#edev redblue (i=star@ppp174.108-253-207.mtl.mt.videotron.ca) |
| 23:50.58 | *** join/#edev dijenerate (n=dijenera@72.22.150.235) |
| 23:56.14 | *** join/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |