| 00:13.59 | *** join/#elinux TimRiker (n=timr@pdpc/supporter/bronze/TimRiker) |
| 00:13.59 | *** mode/#elinux [+o TimRiker] by ChanServ |
| 00:48.00 | *** join/#elinux linac (n=lin@149.83.66.218.broad.fz.fj.dynamic.163data.com.cn) |
| 05:28.25 | *** join/#elinux blindvt_ (n=bf@M997P014.adsl.highway.telekom.at) |
| 05:30.06 | *** join/#elinux blindvt__ (n=bf@M1001P024.adsl.highway.telekom.at) |
| 05:59.22 | *** join/#elinux lyakh (n=lyakh@dynamic-unidsl-85-197-31-110.westend.de) |
| 08:12.09 | *** join/#elinux darkschneider (n=gab@81-208-36-80.ip.fastwebnet.it) |
| 12:39.37 | *** join/#elinux sjhill (n=sjhill@real.realitydiluted.com) |
| 13:29.53 | *** join/#elinux blindvt__ (n=bf@M938P019.adsl.highway.telekom.at) |
| 14:06.31 | *** join/#elinux GPSFan (n=kenm@12.10.255.248) |
| 14:17.05 | *** join/#elinux wmat (n=wmat@CPE000f6641f21a-CM00111adf5f4a.cpe.net.cable.rogers.com) |
| 15:40.09 | *** join/#elinux pprice (n=no@adsl-067-033-058-083.sip.hsv.bellsouth.net) |
| 15:58.08 | *** join/#elinux CosmicPenguin (i=nobody@nat/amd/x-425d538072e17d90) |
| 15:58.08 | *** mode/#elinux [+v CosmicPenguin] by ChanServ |
| 16:00.48 | *** join/#elinux tuxmaniac (n=aanjhan@unaffiliated/tuxmaniac) |
| 16:17.44 | *** join/#elinux ico2 (n=ico2@modem-3214.cheetah.dialup.pol.co.uk) |
| 16:19.41 | *** join/#elinux TimRiker (n=timr@pdpc/supporter/bronze/TimRiker) |
| 16:19.41 | *** mode/#elinux [+o TimRiker] by ChanServ |
| 16:27.18 | cbrake | I'm seeing very slow writes to a USB flash disk from an ARM Linux system. Reads seem to be about what I would expect 1MB/sec. |
| 16:27.41 | cbrake | Any ideas for why copying a 60MB file takes 15 minutes? |
| 16:27.53 | cbrake | Takes 1 min on a Linux PC. |
| 16:31.41 | *** join/#elinux Gerrath (n=Shane_@unaffiliated/gerrath) |
| 16:48.35 | sjhill | cbrake: sounds like your device detects as USB1.1 on your ARM board and USB2.0 on your PC |
| 16:59.20 | davidc__ | cbrake: or your arm board only has a usb1.1 host controller |
| 17:03.29 | cbrake | sjhill: davidc__: yes, ARM is only 1.1, but I'm using an old USB 1.1 flash stick so that I'm running 12MB on both the PC and ARM system. |
| 17:03.45 | davidc__ | cbrake: you mean 12Mbit |
| 17:03.54 | cbrake | sjhill: davidc__: also, my PC test was run on an old PC that does not support USB 1.2 |
| 17:03.58 | cbrake | davidc__: correct 12Mbit |
| 17:04.21 | davidc__ | ah, right. misread your post |
| 17:04.42 | cbrake | davidc__: I'm firing up another PXA270 system just as a sanity check ... |
| 17:04.55 | davidc__ | I read 1Mb/sec as the speed you wanted to improve, not the write speeds. Shows what I get for talking about USB just after getting up |
| 17:05.11 | cbrake | davidc__: :-) no problem |
| 17:05.48 | cbrake | davidc__: 44KB/s write speed is just not acceptible to my customer for some reason |
| 17:05.57 | davidc__ | cbrake: heh, I can understand that |
| 17:06.12 | davidc__ | cbrake: hrm, is it all usb devices that have slow downstream, or just usb flash? |
| 17:06.35 | cbrake | davidc__: I'm not sure. |
| 17:06.42 | TimRiker | mounted without "sync" in both cases? |
| 17:06.51 | davidc__ | cbrake: and have you made sure its not being detected as a low speed [1.5mbit] device? |
| 17:07.12 | cbrake | davidc__: reads are as expected ~1MB/sec. |
| 17:07.19 | cbrake | TimRiker: not sure -- I'll check. |
| 17:07.25 | davidc__ | cbrake: right, right... I'm still only half awake :P |
| 17:07.27 | davidc__ | sorry :P |
| 17:07.30 | TimRiker | I'd try a usb harddrive just for kicks. |
| 17:07.43 | davidc__ | hrm... odd. Ive never seen that before. |
| 17:08.23 | TimRiker | grep Spd /proc/bus/usb/devices |
| 17:08.48 | davidc__ | cbrake: you could stick a logic analyzer on the bus and see whats happening with the bus transactions |
| 17:08.58 | davidc__ | or setup some sort of usb snoop system to see where the delay is |
| 17:09.13 | davidc__ | I've run into plenty of weird USB problems before, but never that... |
| 17:09.40 | sjhill | wow |
| 17:09.46 | sjhill | a Tim Riker |
| 17:09.54 | sjhill | wasn't sure those existed |
| 17:09.56 | TimRiker | ~lart sjhill |
| 17:10.06 | sjhill | heh |
| 17:10.16 | TimRiker | how ya been? |
| 17:10.25 | sjhill | spending your tax dollars like water |
| 17:10.36 | sjhill | pretty good |
| 17:10.37 | sjhill | busy |
| 17:10.44 | sjhill | 2nd child is due at the end of May |
| 17:10.48 | cbrake | TimRiker: yup, its mounted w/ sync: /dev/usbdisk on /media/usbdisk type vfat (rw,sync,fmask |
| 17:10.51 | TimRiker | I'm trying to get more involved in openmoko and similar phone projects, but as always I have a ton of other stuff keeping me busy. |
| 17:10.53 | sjhill | doing a lot of preparations for that |
| 17:12.09 | TimRiker | sjhill: and you think you don't have free time now... :) |
| 17:12.28 | TimRiker | cbrake: sync is safer, but much slower. |
| 17:12.52 | TimRiker | if you go without sync, then be sure to umount before removing. |
| 17:13.09 | cbrake | TimRiker: yes, it looks like udev mounts things w/ sync by default. |
| 17:14.02 | TimRiker | sjhill: I did finally get around to hacking my old xbox. that was fun. then I messed up the flash on the mod board, and had to unhack it. not gotten around to fixing the flash yet. |
| 17:14.17 | cbrake | TimRiker: yes, that was it -- thanks! |
| 17:14.24 | TimRiker | :) |
| 17:14.35 | TimRiker | remember to umount before removing! |
| 17:15.59 | cbrake | TimRiker: yes. |
| 17:17.38 | TimRiker | there is a place in /etc/udev somewhere to change that default. I don't recall where. try grep -Irs sync /etc/udev and see if you can find it. |
| 17:18.52 | *** join/#elinux Gerrath (n=Shane_@unaffiliated/gerrath) |
| 17:22.42 | *** join/#elinux ico2_ (n=ico2@modem-1078.chimpanzee.dialup.pol.co.uk) |
| 17:25.09 | sjhill | cbrake: the usbmount package is also invaluable |
| 17:25.12 | *** join/#elinux pprice (n=no@adsl-067-033-058-083.sip.hsv.bellsouth.net) |
| 17:25.52 | sjhill | TimRiker: well you've been having more fun hacking than i have |
| 17:26.03 | sjhill | i'm still trying to find time to work with my FPGA development board |
| 17:27.10 | kergoth | hey guys |
| 17:27.28 | sjhill | kergoth! |
| 17:27.56 | kergoth | how goes it? |
| 17:47.19 | kergoth | havent had a chance, been busy switching jobs (sorta) once again |
| 17:48.37 | kergoth | guess what i'm doing in the new job... the same shit that drove me nuts in the position in omap at ti. being a driver monkey, churning them out day after day |
| 17:50.59 | *** join/#elinux lyakh (n=lyakh@dynamic-unidsl-85-197-31-110.westend.de) |
| 17:55.17 | wmat | TimRiker: hey there |
| 17:55.26 | wmat | TimRiker: just to let you know, the site merge is moving forward |
| 17:55.39 | wmat | TimRiker: I've been swamped for a month and a half :( |
| 17:56.54 | sjhill | kergoth: there's always truck driving school |
| 17:57.23 | TimRiker | wmat: :) I figured. there have been some updates on elinux.org in the mean time, so we'll have some edits to copy over. |
| 17:58.19 | kergoth | sjhill: hehe. some days i'm tempted to pull an office space and go into construction |
| 17:59.25 | TimRiker | kergoth: There was a hacker at Lineo name pierre that gave it up and decided to play billiards instead of coding. didn't get back online for like 2 years. |
| 18:01.07 | wmat | TimRiker: a friend of mine is writing a new MoinMoin -> Mediawiki converter, which will speed up the basic moving of the content |
| 18:12.57 | sjhill | kergoth: You and I could invent a SAPOTI device? |
| 18:13.41 | sjhill | kergoth: Stab A Person Over The Internet device |
| 18:13.56 | cbrake | sjhill: is there any advantage of usbmount over udev? |
| 18:14.17 | sjhill | usbmount is called by udev |
| 18:14.44 | sjhill | when udev gets the add/remove events, it can then have usbmount properly mount the filesystem with the options you desire |
| 18:15.00 | sjhill | easier than going through all of the hal/dbus madness |
| 18:15.17 | sjhill | i mean, i'm all for chaos sometimes....just not with usb devices |
| 18:17.15 | cbrake | sjhill: :-) |
| 18:18.57 | cbrake | sjhill: looks like on my system, udev is calling a script called /etc/udev/scripts/mount.sh, so I'll probably just tweak that script. |
| 18:27.02 | sjhill | sure |
| 18:29.41 | *** join/#elinux angom (n=angom@red-corp-201.143.88.126.telnor.net) |
| 18:50.37 | *** join/#elinux tonyb2006 (n=tonyb@ool-45712c93.dyn.optonline.net) |
| 18:51.00 | tonyb2006 | <PROTECTED> |
| 18:51.09 | tonyb2006 | when I try to compile a toolchain |
| 19:23.01 | tonyb2006 | well I got that |
| 19:23.19 | tonyb2006 | now |
| 19:23.20 | tonyb2006 | # ./zipitgfx_sample |
| 19:23.20 | tonyb2006 | Error: failed to map framebuffer device to memory. |
| 19:23.24 | *** join/#elinux ico2__ (n=ico2@modem-536.cheetah.dialup.pol.co.uk) |
| 20:51.29 | *** join/#elinux chouimat (n=dieu@kde/developer/chouinard) |
| 21:23.54 | *** join/#elinux ico2_ (n=ico2@modem-2453.chameleon.dialup.pol.co.uk) |
| 21:30.08 | *** join/#elinux blindvt_ (n=bf@M940P000.adsl.highway.telekom.at) |
| 22:01.44 | *** join/#elinux TimRiker (n=timr@rikers.org) |
| 22:01.45 | *** mode/#elinux [+o TimRiker] by ChanServ |
| 22:07.07 | *** join/#elinux chouimat (n=dieu@kde/developer/chouinard) |
| 23:24.35 | *** join/#elinux ico2__ (n=ico2@modem-753.chimpanzee.dialup.pol.co.uk) |